Feature Quantification and Abnormal Detection on Cervical Squamous Epithelial Cells

Table 1

Abnormal cell detection based on AP algorithm with four indicators: the size of nucleus, N/C, circularity, and compactness.

(a)

Clustering	Distance of samples
	Criterion: the size of nucleus
Categories	I	II	III	IV	V	VI
Number of clustering centers	33	8	31	20	27	23
Samples in the class	5, 6, 19, 22	1, 13	4, 7, 10, 12, 14, 18, 21, 28	2, 9, 11, 15, 16, 17, 26, 32	3, 24, 25, 34	29, 30
Scope of feature values	(1.1, 1.4)	(1.7, 1.9)	(2.2, 2.9)	(3.4, 3.9)	(4.0, 4.5)	(5.1, 6.2)
Feature thresholds

(b)

Clustering	Distance of samples
	Criterion: N/C
Categories	I	II	III	IV	V	VI
Number of clustering centers	17	5	8	26	12	25
Samples in the class	18, 19, 20, 21, 22, 31, 33	15, 32	4, 6, 27, 30	1, 2, 3, 9, 11, 14, 16	13, 24, 29	7, 10, 23, 28, 34
Scope of feature values	(1.7, 4.7)	(8.3, 9.3)	(11.9, 15.2)	(16.5, 19.9)	(21.6, 25.5)	(26.5, 31.1)
Feature thresholds

(c)

Clustering	Distance of samples
	Criterion: circularity
Categories	I	II	III	IV
Number of clustering centers	5	2	24	19
Samples in the class	1, 3, 4, 6, 8, 9, 11, 12, 13, 15, 16, 18, 20, 21, 22, 26, 28	17, 29, 30, 32	7, 10, 14, 25, 27, 33, 34	23, 31
Scope of feature values	[0.8, 0.9]	[0.8, 0.8]	[0.8, 0.8]	[0.5, 0.7]
Feature thresholds

(d)

Clustering	Distance of samples
	Criterion: compactness
Categories	I	II	III	IV
Number of clustering centers	33	29	23	26
Samples in the class	4, 7, 15, 19, 30, 31, 32	1, 6, 17, 28	8, 11, 14, 18, 20, 27, 34	2, 3, 12, 22, 24
Scope of feature values	[0.5, 0.7]	[0.8, 0.8]	[0.8, 0.9]	[0.9, 0.9]
Feature thresholds